An Approach to Designing Software Engineering Thesis Papers

D. Koznov and M. Nemeshev

Software Engineering Department, Saint Petersburg State University, Saint Petersburg, Russia

Keywords: E-learning, Writing Theses, Mind Maps, Software Engineering.

Abstract: Thesis development proves most complicated for students, as it implies more than one final work product and

involves facing deadlines. One of the most challenging aspects here is writing the thesis paper. Software

engineering is marked by specific difficulties, since students focus on the software development and neglect

other aspects, first of all, the thesis paper writing. Hence thesis paper writing demands significant support

from the university. This paper presents a method of managing the thesis paper writing process for software

engineering undergraduates. Mind mapping is used for the early design, fast feedback, and a wide discussion

of the thesis paper plans. The method is implemented as a one semester course for last year undergraduates.

The evaluation conducted indicates a significant increase in the quality of thesis papers.

1 INTRODUCTION

Preparing a thesis is no less challenging a task for

students, than for organizations who instruct and tutor

thesis writing at higher education institutions (Rapp,

Kruse, 2016). Following (Davis, Parker, 1979), the

thesis writing process takes one third of the total time

allocated to obtaining the degree; and although these

authors were discussing the doctoral thesis, the

statement is also relevant for other types of theses.

Most students, however, tend to underestimate the

effort that a thesis demands (LaCourse, Rock, 2002).

Hence, they need a considerable educational,

resource and management support to write their

theses properly (Aghaeea, Keller, 2016). Providing

such support for undergraduates is especially

challenging for a number of reasons. Firstly,

undergraduates have fewer professional and

academic skills than other kind of students, and most

of them are working on the first thesis in their life.

Secondly, they by far outnumber graduate and

postgraduate students, so it is quite difficult to

provide high quality one-to-one individual support.

Software engineering involves an additional

difficulty, while students have a lot of software

development to do, and neglect preparing other work

products, mainly thesis papers.

At our department, every year 30 to 40

undergraduates face the task of preparing a thesis.

Besides individual supervision, we provide an

additional kind of centralized support. Managing

writing thesis proves to be one of the most significant

parts of student work that needs to be supported.

There is a number of approaches to providing

good quality thesis papers (Rapp, Kruse, 2016),

(LaCourse, Rock, 2002), (Aghaeea, Keller, 2016),

etc. However, there is a lack of approaches which

would work in conjunction with the actual thesis

development. Also, some additional techniques need

to be used in the thesis development process to ensure

collaboration and to encourage students.

Mind mapping (Buzan, 1995) is a widely known

technique to structure and manage any kind of

information. It is actively used in education, solving

a wide range of tasks: supporting collaboration,

stimulating critical thinking, improving memory

power, etc. (Fun, Maskat, 2010), (Noonan, 2013),

(Papushina, et al., 2017). However, this technique has

not been used so far for the management of thesis

writing at the department level.

The present paper describes a method for the

management of thesis writing for software

engineering undergraduates. The aim of the method

is to support thesis design, to provide collaboration

between the author, the supervisor and other

department staff, to ensure fast feedback at an early

stage of writing. The method is based on the mind

mapping technique and the Comapping toolset

(Comapping, 2017). It is implemented as a course for

last year undergraduates. During the course, students

develop plans of their thesis papers through following

suggested patterns and guidelines. Students discuss

Koznov D. and Nemeshev M.

An Approach to Designing Software Engineering Thesis Papers.

DOI: 10.5220/0006497301390145

In Proceedings of the 9th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (KMIS 2017), pages 139-145

ISBN: 978-989-758-273-8

these plans with their supervisors and the course

lecturer, and substantially modify them. Finally,

every student prepares two or three drafts of his/her

thesis paper, following the plan developed before. By

teaching this course, the department closely monitors

students’ progress at an early stage of thesis writing.

Further work is managed by student supervisors

alone. The department staff then check the final

versions of the theses. Our paper describes the

method and course structure, and also presents some

evaluation data.

2 RELATED WORK

There are a large number of books and papers on

academic writing, e.g. (Annesley, 2010), (Swales,

Feak, 2004), (Canagarajah, 2013). In (Stojmenovic,

2010) specific guidelines for writing papers on

software engineering and computer science are

presented.

There exists significant literature on development

software documentation: see (Barker, 2002) for

survey tools and methods. There is also a variety of

approaches to writing user documentation (Weiss,

1991) and a vast literature on technical writing

(Williams, 2000), (Wright, 2010).

However, academic and technical writing courses

can help students in preparing their thesis papers only

when their general writing skills are already high

enough, which, in fact, does not prove to be the case

even after taking these courses (LaCourse, Rock,

2002). Hence a special form of thesis development

support is required.

There are a number approaches, techniques, and

tools to support thesis development.

The problem of interaction between students and

supervisors during thesis writing is considered in

(Hansen, Hansson, 2015). The authors analyze web-

discussions between students and supervisors, and

reveal that one of the main topics of collaboration was

«How to write».

Peer reviews to improve the quality of thesis

papers are considered in (Aghaeea, Keller, 2016),

(Aghaee, Hansson, 2013). The findings show a

considerable investment of the approach to increasing

thesis paper quality. Also, some limitations of the

approach are reported: the motivation of students to

write reviews, student review skills, the quality of

papers to be reviewed. To apply the approach

efficiently, thesis papers should appear as readable

and well-understandable drafts quite early. However,

our experience shows that working on the thesis paper

is often the last thing that undergraduates do in their

thesis projects, which leaves very little time to review

and discuss the texts.

The necessity of a special course that is directly

connected with the actual thesis project is argued in

(LaCourse, Rock, 2002). An early 1-credit course for

graduate students is offered. The course includes the

following topics: time management, scientific

method, thesis statement, justification and

background, the structure of the thesis paper, and

some other issues. The course aims to support the

student’s transition to an independent researcher.

However, the course addresses to graduate students,

who need to master individual research skills.

Meanwhile software engineering undergraduates are

less mature and academically oriented and focus more

on acquiring basic professional skills.

A number of special software systems are

developed to support thesis development: Thesis

Writer (Rapp, Kruse, 2016), SciPro system (Aghaeea,

Keller, 2016). These systems provide thesis

guidelines, support collaboration between students

and supervisors, provide facilities for peer-reviews,

and allow to collect various statistics to improve the

process. Such kind of software is useful for managing

thesis development, yet it cannot fully replace special

courses. Moreover, additional techniques need to be

employed to encourage students to put more effort

into writing and to facilitate thesis development

process.

A promising technique for structuring and

analyzing information that is actively applied in

education is mind mapping (Buzan, 1995).

Teacher-centered and student-centered mind

mapping modes are explored in (Fun, Maskat, 2010).

The authors found that the active involvement of

students in mind mapping is far more efficient than

presenting them with high quality maps constructed

by teachers.

Educational modules within the technical

secondary school essentially using mind mapping are

presented in (Tee, Yunos, 2012). Students use mind

mapping for reading literature, revising, note-taking,

etc. The study found the technique improves students’

learning achievements.

There exists numerous evidence on using mind

mapping in health education. In (Edwards, Cooper,

2010) mind mapping is considered as a teaching

resource in clinical education. It is explained how to

use this technique for preparing and reviewing

lectures, and for examinations. To address the issues

of poor paper quality, a faculty program is offered in

(Bickes, Schim, 2010). The program includes a

writing workshop, a revision of the grading rubric,

and a system of blind review for grading student

papers. However, the program is provided mostly

prior to rather than in conjunction with the actual

thesis development. An interesting survey of mind

mapping in health education is presented in (Noonan,

2013). 11 papers were analyzed, and some of them

reported unsuccessful results of mind mapping. It

seems that one of the crucial points for successful

mind mapping is the students’ openness and

willingness to try new techniques. Otherwise, it is

quite difficult to shift the situation.

(Somers, et al., 2014) and (Papushina, et al., 2017)

report about applying mind mapping in business

education. The former paper presents the results of an

experiment, where undergraduate and MBA students,

and faculty staff developed mind map models for the

same complex topics. This revealed some interesting

facts concerning students’ abilities to integrate and

apply knowledge. For example, the students

represented their knowledge as a set of poorly

connected «islands»; in contrast, the faculty staff

developed densely connected models for the same

topics. The latter paper discusses mind mapping

software tools for education needs, and also presents

the results of using mind mapping as a part of a

Master’s program in marketing. The students use

mind mapping in outclass reading and to present the

results of their research tasks, their mind maps being

then discussed and evaluated in class.

Computer-supported collaborative mind mapping

in software engineering education is considered in

(Koznov, Pliskin, 2008), (Koznov, 2012). Mind

mapping is used as an exam technique for the final

software engineering course and for designing term

papers and theses.

(Kokotovich, 2008) considers non-hierarchical

mind mapping in industrial design. The author offers

this technique for students and novice designers in the

early phases of the problem-solving process. Non-

hierarchical mind map means a network data

structure.

Summarizing, there is no research on applying

mind mapping for thesis development in the context

of a special course. This idea is discussed briefly in

(Koznov, 2012), but no integrated method has been

offered so far.

3 MIND MAPS AND COMAPPING

Mind mapping was suggested by Tony Buzan in the

1970s to work more efficiently with any kind of

information (Buzan, 1995). The idea is to use a simple

diagramming notation: the primary (central) object is

put in the middle, secondary objects, which clarify the

meaning of the central one, are drawn around it and

connected with it, etc. (Fig. 1).

Figure 1: An example of a mind map.

This approach is suitable for analyzing and

understanding any kind of information, especially in

case of large volumes. The approach is widely used

in education, business, psychology, and other

domains. There are also a number of software tools

implementing the approach. Further details can be

found in (Koznov, Pliskin, 2008), (Koznov, 2012),

(Papushina, et al., 2017).

Comapping (Comapping, 2017) is a collaborative

online mind mapping tool. It supports features such

as easy drag and drop, smooth animation, large maps

with smart auto-focusing features, etc. A tree-like

notation (left-to-right mind mapping) is better for the

computer-based support when combined with a

layout algorithm, as it is easier to read and understand

than the centre-based one. An example of a mind map

in Comapping is shown in Fig. 2.

The tool allows sharing maps among any number

of users, as well as notifying them about map

changes. These changes are highlighted on the map

with detailed information, and other users can review

the changes and leave comments, making Comapping

a suitable tool for professors to monitor student work.

4 METHOD

The main idea of the method is to use mind mapping

for the early design of student thesis papers. It enables

collaboration between department staff, thesis

supervisors and students to develop a mature thesis

plans following the patterns and guidelines

recommended by the department. Our experience

shows that students find it extremely difficult to make

serious corrections to the paper when the bulk of it is

already written. On the other hand, reading and

correcting a poorly written text is far too laborious for

the supervisor and the department staff. The goal can

be achieved more efficiently at the paper design

phase. Just as visual models in software engineering

(UML, SysML, etc.), and in business modeling

(BPMN, EPC, SADT, etc.), mind mapping is able to

provide easy information exchange and fast feedback.

Figure 2: Example of a mind map diagram in Comapping.

The first steps of the method are defining thesis

goals, specifying the results, and creating paper

backbone (see fig. 3). The result of these steps is the

mind map of a student’s thesis paper. Then the

introduction is planned and written, and the first two

or three drafts of the paper are developed. Let us

consider every step in more detail.

Recovering

thesisgoals

Clarifyingresult

objects

Creatingpaper

backbone

Planningand

writing

introduction

Writing&discussing

firstpaperdrafts

Figure 3: Method schema.

Defining thesis goals aims to reconsider and

renew the goals of the thesis. Our method is applied

when the practical part of the thesis project is done,

and the student starts to write the thesis paper. As our

experience shows, at this stage he/she very often fails

to give a clear explanation of the thesis goals, with

only the supervisor being able to explain the choice

of the topic, the use of particular technologies, etc.

Undergraduates tend to view the thesis statement as a

set of requirements to fulfill, instead of an issue to

contemplate and discuss. The first step is thus

intended to raise their awareness of the thesis goals.

When they start writing their text seems the most

suitable time to reconsider this information. We

recommend compiling a list of thesis goals.

Specifying the results. It is not very easy for

undergraduates to identify the results of their work.

What exactly has been done? An algorithm, software

architecture, software, experiments? It does not mean

the student has produced nothing. Often it means they

need help to identify various work products.

Sometimes, some of these products require additional

work, such as appropriate testing and evaluation of

the software developed. All these work products will

be thesis results, and it takes some effort and time to

specify them. Ideally, thesis results should be

presented as a list, with each item of the list having a

brief description of the corresponding work product.

We normally recommend our students to include this

list in the conclusions section. It is also important that

the goals correlate to the results.

Creating paper backbone. The main purpose of

the thesis paper is to present and describe the results

of the thesis project. At the previous steps, the student

defines the goals and specifies the results of the

thesis. We recommend to map every result (thesis

work product) to a separate chapter of the thesis

paper. The set of chapters and their subsections are

specified on the thesis mind map. Additional chapters

are Background and Related Work, which are also

specified on the mind map. In case of industrial

theses, Related Work means a description of existing

software solutions similar to the one developed by the

student in the thesis project.

Planning and writing the introduction. This is

important as a special step because the introduction

commonly poses a serious challenge to many

students, as it has to include a brief description of

thesis contexts and involved fields, and a clear setting

of the goals. The plan of introduction is developed in

the mind map for an easier discussion with the course

lecturer.

Writing & discussing first paper drafts. It is

important to transfer the mind map correctly into the

first draft of the thesis paper. This has to be assisted

because this is a stage when many errors are spotted

and corrected, and misunderstandings are clarified.

Moreover, students sometimes start writing their

thesis papers neglecting the plan they developed

before. Writing the first drafts helps to avoid this, the

draft serving as a guarantee that the students will

continue their work by following initial developed

plans.

5 COURSE

The method is implemented as a semester course. The

course is mandatory for all last year undergraduates

of our department. It is held in the last semester when

our students work on their theses. The flowchart of

the course is presented in Fig. 4. Let us now consider

the course in more detail.

Lecturesabout

text writing (2-3)

Lectureabout

mindmapping

Comapping

demonstration

Collaborative

process

establishment

Briefexplanation

ofthesis

Mindmapdevelopment

(2‐5iterations)

Th esis paper

development

(2‐3iterations)

Collaborative work

Lectures

Figure 4: Flowchart of the course.

Lectures about text writing. These lectures ensure

that students understand the role of text in thesis

development and become familiar with the

recommended patterns and guidelines.

Lecture about mind mapping. Students learn

about mind mapping and other types of information

visualization techniques (Kudryavtsev, Gavrilova,

Leshcheva, 2013), about knowledge management

(Gavrilova, 2010) and the features of mind mapping.

Comapping demonstration. Since this software is

simple and ergonomic, a demonstration of its main

features is sufficient for the students to learn to use it.

Collaboration process establishment. The process

of collaboration is explained and agreed on with the

students. They are also given access to Comapping.

The next steps are conducted as one-to-one

sessions of the course lecturer and the course student.

One or two members of the department staff help the

course lecturer. It should be noted that mind mapping

allows to spend 10-15 minutes per student to verify

and discuss the current version of the thesis plan.

Often the lecturer and his/her assistants verify mind

maps before class to save time.

Brief explanation of thesis. The course

lecturer/assistant is not supposed to go deeply into the

details of each thesis because it would take too much

effort and time. However, he/she needs to have a clear

idea of each thesis in order to be able to provide

appropriate recommendations. With most theses, it

does not pose much difficulty as they are supervised

by the department’s industrial partners or by the

department staff, so the course lecturer is already

familiar with the topic areas. There can be a few

theses on topics from new industrial companies or

from staff from other university departments, in that

case understanding their ideas may be less easy for

the course lecturer.

Mind map development. Every course student

creates a plan for his/her paper under the course

lecturer’s/assistant’s supervision. It usually takes 2 to

5 iterations, depending on how intensively the student

works. Ideally, after the first iteration, the student

discusses the plan with his/her supervisor and then

corrects it.

Thesis paper development. Two or three iterations

are optimal for the purpose. The first iteration

normally reveals certain problems (for example, the

plan turns out to be inadequate and the student has to

go a step back to improve it). The following one or

two iterations give the students a chance to correct the

found errors. This step is the most labor intensive for

the course lecturer and assistants, because they need

to read a large number of papers. However, it should

be noted that these papers contain only the

introduction (with thesis goals), conclusions (with

results), and the thesis contents page. Thus, it is quite

easy to read them. It is a challenge, though, to

organize the course work, i.e. to avoid the situation

when most students submit their papers at the same

time at the end of the course, which causes the course

lecturer and assistants to overwork.

6 EVALUATION

We have delivered this course for the past 7 years,

with more than 300 students having taken it so far.

We have studied the department’s thesis papers of the

last 10 years and witnessed a considerable

improvement in their quality. We believe that the

course presented here made a significant contribution

to this progress.

In addition, we conducted an evaluation of the

method and the course. We prepared a sample

selection of papers, developed both with and without

the course assistance. We then asked several

independent experts, who did not know about the

course, to estimate the quality of the papers. Our

hypothesis was that the course assisted papers would

be found by the experts to be of a higher quality.

Below we describe the evaluation in more detail.

We selected 4 papers which had been assisted by

the course and another 4 papers which had been

written without the course assistance. For every paper

we removed all the information about the author and

the supervisor.

We developed a questionnaire for experts to

evaluate paper quality. The questionnaire contains the

following questions:

1. What is your level of competence in the

subject area?

2. How would you characterize the degree of

the paper’s complexity?

3. Is the topic easy to understand?

4. Can the thesis goals be easily understood?

5. Are the motivation and relevance of the paper

clear?

6. Is it easy to see what exactly has been done

by the student (i.e. the results)?

7. Is it possible to judge how profound the

results are?

8. How comprehensive is the description of the

results?

9. What degree of text integrity does the paper

demonstrate?

10. Is the paper free from unnecessary details?

11. Does the structure of the paper help to gain

an insight into the matter?

12. Is it written in good Russian/English?

13. Is text formatting good enough?

14. How appropriate is the use of special

terminology?

15. Would you recommend this text to anyone?

We used a five-point Likert-type scale which

ranges from “excellent” to “very bad”.

We found 10 external experts to read and evaluate

the papers. Six experts had an industrial background

(managers and software developers), while four had

an academic background. The results of the survey

are presented in Table 1.

Table 1: Evaluation results.

Course assisted course unassisted

4,8 3,3

4,7 3,6

4,6 3,8

4,7 4,1

4,7 3,7

The first column contains the total scores for

thesis papers that had been assisted by the course. The

second column contains the total scores for the

unassisted papers. For every paper, the total score was

calculated as the average of scores for each question

(except the first two). The bottom row of the table

shows the average scores for course assisted and

course unassisted papers.

The data in Table 1 confirms our hypothesis:

course assisted papers scored the average of 4.7,

while course unassisted papers – only 3.7.

It should be noted that the industrial background

experts tended to generally like or dislike the papers

rather than to be scrupulous, and evaluated the papers

accordingly. This resulted in a wide scatter in the

grades they gave. The representatives of the academia

were more precise and rigorous in the assessment. A

more extensive evaluation remains a task for the

future.

7 CONCLUSIONS

The current paper presents a method for the

management of thesis writing for software

engineering undergraduates. The method is based on

mind mapping for the early design and discussion of

the thesis paper plan. It is implemented as a one

semester course for last year undergraduates. The

evaluation conducted indicates a significant increase

in the quality of thesis papers.

One aspect for further research could be the use of

thesis mind maps for peer student reviews or other

kinds of reviews. It can prove beneficial as thesis

mind maps appear much earlier than papers. Also,

mind maps are easier to review, and corrections can

be done faster.

It should be noted that the method and the course

are suitable for both undergraduate and graduate

students, and can be applied not only in software

engineering, but also in many other education fields.

The method can be modified depending on the

particular requirements. Graduate students, for

example, need a more academic, research-based

focus. Other education fields may offer various other

thesis paper templates. What is important is that the

core of the method suits any student specialization.

Some preliminary steps to adapt the method to

courses in design and business education fields have

already been taken.

REFERENCES

Rapp, Ch., Kruse, O., 2016. Thesis Writer (TW): Tapping

Scale Effects in Academic Writing

Instruction. L@S 2016.

Davis, G., Parker, C., 1979. Writing the Doctoral

Dissertation, Woodbury, NY: Barron's Educational

Series.

LaCourse, P., Rock, B., 2002. A course initiating an

engineering thesis. ASEE Annual Conference

Proceedings.

Aghaeea, N., Keller, C., 2016. ICT-supported peer

interaction among learners in Bachelor's and Master's

thesis courses. Computers & Education, 94.

Aghaee, N., Hansson, H., 2013. Peer portal: Quality

enhancement in thesis writing using self-managed peer

review on a mass scale. International Review of

Research in Open and Distance Learning.

Hansen, P., Hansson, H., 2015. Optimizing student and

supervisor interaction during the SciPro thesis process

– concepts and design. Lecture Notes in Computer

Science, 9412.

Bickes, J., Schim, S., 2010. Righting writing: Strategies for

improving nursing student papers. International

Journal of Nursing Education Scholarship 7 (1), 8.

Annesley, T., 2010. Clinical Chemistry, 56 (3), 56(4),

56(5), 56(6), 56(7).

Swales J., Feak C., 2004. Academic writing for graduate

students: Essential tasks and skills. Vol. 1. Ann Arbor,

MI: University of Michigan Press.

Canagarajah, A., 2013. Critical academic writing and

multilingual students. University of Michigan Press.

Barker, T., 2002. Writing software documentation: A task-

oriented approach (Part of the Allyn & Bacon Series in

Technical Communication, Longman).

Weiss, E., 1991. How To Write Usable User

Documentation. 2nd edition, Phoenix: Oryx Press.

Stojmenovic, I., 2010. Editor's Note: How to Write

Research Articles in Computing and Engineering

Disciplines. IEEE Transactions on Parallel and

Distributed Systems, 21(2).

Williams, J., 2000. Transformations in technical

communication pedagogy: engineering, writing, and

the ABET engineering criteria. SIGDOC.

Wright, H., 2010. Technical Writing Tools for Engineers

and Scientists. Computing in Science and Engineering,

12(5).

Buzan, T., 1995. The mind map book. 2nd edition, BBC

Books, London.

Fun, C., Maskat, N., 2010. Teacher-centered mind mapping

vs student-centered mind mapping in the teaching of

accounting at pre-U level – an action research. Procedia

Social Behavior Science, 7.

Tee, T., Yunos, J., et al, 2012. The development and

implementation of Buzan mind mapping module.

Procedia Social Behavior Science, 69.

Edwards, S., Cooper, N., 2010. Mind mapping as a teaching

resource. Clinical Teaching, 7(4).

Noonan, M., 2013. Mind maps: enchancing mindwifery

education. Nurse Educ. Today 33.

Somers, M., Passerini, K., Parhankangas, A., Casal, J.,

2014. Using mind maps to study how business school

students and faculty organize and apply general

business knowledge. Journal of Management

Education, 12 (1).

Papushina, I., Maksimenkova, O., Kolomiets, A., 2017.

Digital educational mind maps: A computer supported

collaborative learning practice on marketing master

program, Advances in Intelligent Systems and

Computing, 544.

Koznov, D., Pliskin, M., 2008. Computer-Supported

Collaborative Learning with Mind-Maps. T. Margaria

and B. Steffen (Eds.): ISoLA 2008, CCIS, 17.

Koznov, D., 2012. Teaching to Write Software Engineering

Documents with Focus on Document Design by Means

of Mind Maps. International Conference on Computers

and Advanced Technology in Education (CATE2012).

Kokotovich, V., 2008. Problem analysis and thinking tools:

an empirical analysis of nonhierarchical mind

mapping. Design Studies. 29 (1).

Comapping, www.comapping.com (accessed May 5,

2017).

Kudryavtsev, D., Gavrilova, T., Leshcheva, I., 2013. One

approach to the classification of business knowledge

diagrams: Practical view. Federated Conference on

Computer Science and Information Systems (FedCSIS).

Gavrilova, T., 2010. Knowledge engineering for non-

engineers. 6th IFIP International Conference on

Intelligent Information Processing, IIP.